Nowadays, many software projects are partially or completely open-source based. There is an increasing need for companies to participate in open-source software (OSS) projects, e.g., in order to benefit from open source ecosystems. OSS projects introduce particular challenges that have to be understood in order to gain the benefits. One such challenge is getting newcom- ers onboard into the projects effectively. Similar challenges may be present in other self-organised, virtual team environments. In this paper we present preliminary observations and results of in-progress research that studies the process of onboarding into virtual OSS teams. The study is based on a program created and conceived at Stanford University in conjunction with Facebook’s Education Modernization program. It involves the collaboration of more than a dozen international universities and nine open source projects. More than 120 students participated in 2013. The students have been introduced to and supported by mentors experienced in the participating OSS projects. Our findings indicate that mentoring is an important factor for effective onboarding in OSS projects, promoting cohesion within distributed teams and maintaining an appropriate pace.

1. Onboarding in Open Source Software Projects:
A Preliminary Analysis
Jürgen Münch∗
with Fabian Fagerholm∗
, Patrik Johnson∗
,
Alejandro Sanchez Guinea∗
, Jay Borenstein†
∗
University of Helsinki
†
Stanford University
†
Facebook
26 August 2013
VirtuES Workshop (colocated with ICGSE’11)

2. Slide 2
Overview
 Motivation
 Context
 Research Approach
 Results
 Limitations
 Conclusions and Future work

3. Slide 3
Motivation
 Most companies need to develop their software with
so-called software ecosystems
 reasons: software size exceeds capabilities of an individual
organization, need to combine own offerings with
infrastructure(s) of ecosystem, market dominance of specific
ecosystem, ...
 OSS projects (such as Android, LAMP stack) play a
major role as software ecosystems
 similar to Global Software Development (GSD) in many ways
 projects are often highly decentralised
 with participants from a wide range of geographical locations
and cultural backgrounds
 But the development process followed in many OSS
projects is specific
 Developers have a large degree of freedom to decide how to
manage themselves within their projects

4. Slide 4
Motivation
 Companies need to get involved in OSS projects quickly
 how to support the entry of new members into OSS projects?
 guidance to address this is widely missing
 OSS development teams are typically self-organized
virtual teams
 lack of formally appointed leaders or indications of rank or role
 shared power
 self-assignment
 “soft delegation”
 Onboarding or Organizational Socialization
 “the process that helps new employees learn the knowledge,
skills, and behaviours they need to succeed in their new
organizations” [1]
 potential impact factors on the effectiveness and efficiency of
onboarding: mentoring, project type, face-to-face meetings, ...
 Question: What is the impact of mentoring in OSS projects?

5. Slide 5
Context
 Facebook‘s Education Modernization program 2013
 9 open source projects
 130 students, > 12 Universities
 Developers submitted their OSS project preferences
 Based in this they were assigned to virtual teams with 4–8
participants from at least two different universities
 Each OSS project had two separate virtual teams working
for it, and provided an experienced senior developer to
mentor the developer teams
 Onboarding support: kickoff and continuous mentoring in the
project
 Focus of this study: 4 OSS projects with participation of
University of Helsinki through the Software Factory
Laboratory

6. Development in the Software Factory
at University of Helsinki
Slide 6

7. Slide 7
Context

8. Slide 8
Kickoff
 Three-day kickoff event at Facebook‘s Menlo Park
headquarters
 Students got acquainted with other members of their virtual
teams and their open source mentors
 Mentors provided practical training

9. Slide 9
Research Approach
 Goal:
 characterise and understand the onboarding process in the
OSS projects under study, with particular consideration of the
role of mentoring as part of that process
 ultimate aim is to derive guidelines for onboarding. e.g., how to
design a mentoring program
 Focus:
 contribution level over time, i.e., productivity and
communication activity among team members
 Design:
 quantitative multiple-case study approach
 compare a subset of the developers receiving mentoring
support against developers in the same OSS projects who did
not receive the mentoring and enter OSS projects in the “usual
way“
 Sample:
 random sample of 20 developers for each group
 first 12 weeks of the project

10. Slide 10
GQM Plan
 ...

11. Slide 11
Measures
 Aggregate Measure „Activity“
 Activity = #commits + #pull_requests + #interactions
(inspired by Fritz et al. [2])
 Commit
 submission of the latest changes in the source code to the
repository in the context of revision control systems
 Pull Requests
 a notification that is made to others about changes pushed to
the repository, so that interested parties can review changes,
discuss potential modifications and push follow-up commits if
necessary (as characterized by GitHub) [3]
 Interaction
 single message posted in a GitHub discussion forum

12. Slide 12
Preliminary Results
Developers in the mentored group show roughly three times
more activity than the non-mentored group

13. Slide 13
Preliminary Results
 ...
same level
Three-week
plateau
steady increase
Significant increase

14. Slide 14
GQM Plan
 ...
with
mentoring
more
interactions
more
time spent
more commits
and pull requests
-
-
large
influence

15. Slide 15
Results form the Ruby on Rails Project
 Speeding up Ruby on Rails
 Students have been working on the Active Record database
access component
 Active Record is an object-relational mapper that allows for
abstracting an underlying relational database. Thus developers
do not need to write database queries directly.
 The students implemented a statement cache for Active
Record. The statement cache makes Active Record faster by
reusing data structures that would otherwise be re-generated
on each database query.
 The new statement cache has been merged into the release of
Ruby on Rails 4.0, which is a major update to the web
framework.

16. Slide 16
Limitations of this Preliminary Study
 We did not distinguish between mentoring and other
onboarding support factors (such as kickoff event)
 We included only part of the data available (4 of 9 projects,
only first 12 weeks)
 We used student subjects
 We have not examined whether developers in the non-
supported group have received any treatment during the
observed period of time that would impact the results
 We have not statistically evaluated the construct validity of
our activity metric
 The subjects may have altered their behaviour in response
to the study

17. Slide 17
Summary and Future Research
 We thus find support that onboarding support increases the
chance of developers to be exposed to, select, and perform
tasks in a proactive and self-directed manner in open source
projects.
 We assume that this applies in other kinds of GSD settings
where the organisational and team structure is similar.
 Future research
 Expansion of the sample
 Analyzing commits, pull requests, and interactions over time
 Developing more comprehensive measures
 Developing a more comprehensive set of measures
 Taking other impact and context factors into account
(such as project type, mentoring type, etc.)
 We assume that mentoring is highly important for
maintaining developers’ motivation and for attaining good
cohesion within virtual teams

18. Slide 18
Contact and Further Information
@Juergen_Muench
 Software Systems Engineering Research Group
http://www.sserg.org
 Software Factory
http://www.softwarefactory.cc
 Software Factory video
http://www.youtube.com/watch?v=uPAE1uRP65Y
 Software Factory Speeds up Ruby on Rails (article)
http://www.sserg.org/software-factory-speeds-up-ruby-on-
rails/

19. Reference of the Article
Fabian Fagerholm, Patrik Johnson, Alejandro Sánchez Guinea,
Jay Borenstein, Jürgen Münch. Onboarding in Open Source
Software Projects: A Preliminary Analysis. In Proceedings of
the 8th International Conference on Global Software
Engineering (ICGSE 2013), Compendium Proceedings
(VirtuES Workshop), Bari, Italy, August 2013. IEEE.
The article can be found online here:
http://www.sserg.org/publications/uploads/64801075c6f8709a4
9f0514afac282e594c0e271.pdf
Slide 19

20. Further References
[1] T. N. Bauer and B. Erdogan, “Organizational socialization:
The effective onboarding of new employees.” 2011.
[2] T. Fritz, J. Ou, G. C. Murphy, and E. Murphy-Hill, “A degree-
of- knowledge model to capture source code familiarity,” in
Proceedings of the 32nd ACM/IEEE International Conference
on Software Engineering- Volume 1. ACM, 2010, pp. 385–394.
[3] Github, Inc., “Using Pull Requests.” [Online]. Available:
https: //help.github.com/articles/using-pull-requests
Slide 20